Staphylococcus aureus is the single most common cause of bacterial infections among drug users and is increasingly recognized as a major cause of life-threatening disease in patients with AIDS. Both drug users and HIV-infected individuals have an increased incidence of colonization and subsequent infection with S. aureus. Little is known about the biology of S. aureus nasal colonization or the basis for the increased rate of colonization among any of the known high-risk groups. In a community study of active drug users, we found several biologically linked networks of subjects colonized with closely related strains of S. aureus. The biologic linkage of the strains, established by pulsed field gel electrophoresis (PFGE), first identified these networks. This study demonstrated a relationship between the drug use setting, a "crack house", and the biologically linked networks of S. aureus. The overall goal of this proposal is to define the extent, patterns and mechanisms of S. aureus transmission among drug use risk networks. Our specific aims are the following. 1. Establish the baseline distribution of S. aureus colonization in drug use networks. 2. Explore the extent to which the prevalence and strain similarity of S. aureus colonization are associated with the structure of drug use networks, risk behaviors and features of drug use settings. 3. Examine whether co-factors, such as HIV status influences colonization. 4. Determine the contribution of drug use paraphernalia to the transmission of S. aureus. 5. Analyze staphylococcal isolates using PFGE and multilocus sequence typing, in order to define strain identity and the degree of relatedness among isolates, correlate them with international and regional strain databases, supplement social network linkages identified through interview with biologic linkage data and determine which typing tool is most useful in this setting. 6. Use the strain profiles generated by the typing tools to identify virulence determinants that confer an ecologic advantage to these strains, enabling them to colonize, resist antibiotics and spread among subjects. The present study takes advantage of Dr. Miller's preexistent cohort of women drug users and their network members in the Bedford Stuyvesant section of Brooklyn. This will allow for the rapid integration of this investigation into her ongoing study.